Abstract
Evaluation for probabilistic multiclass systems has predominately been done by converting data into binary classes. While effective in quantifying the classifier performance, binary evaluation causes a loss in ability to distinguish between individual classes. We report that the evaluation of multiclass probabilistic classifiers can be quantified by using the area under the distance threshold curve for multiple distance metrics. We construct our classifiers for evaluation with data from the National Cancer Institute (NCI) Lung Image Database Consortium (LIDC) for the semantic characteristic of malignancy. We conclude that the area under the distance threshold curve can provide a measure of the classifier performance when the classifier has more than two classes and probabilistic predictions.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Bradley, A.P.: The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recogn. 30, 1145–1159 (1997), doi:10.1016/j.bbr.2011.03.031
Drummond, C., Holte, R.: Cost curves: an improved method for visualizing classifier performance. Mach. Learn. 65, 95–130 (2006)
Zinovev, D., Furst, J., Raicu, D.: Building an ensemble of probabilistic classifiers for lung nodule interpretation. In: Tenth Intern. Conferen. on Mach. Learn. and App (ICMLA 2011), pp. 151–167. IEEE Press (December 2011), doi:10.1109/ICMLA.2011.44
Amor, N.B., Benferhat, S., Elouedi, Z.: Information-based evaluation functions for probabilistic classifiers. In: Eleventh Internat. Conferen. on Infor. Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU 2006), pp. 428–433 (July 2006)
Ling, C.X., Huang, J., Zhang, H.: AUC: a statistically consistent and more discriminating measure than accuracy. In: Proc. of Eighteenth Internat. Conf. on Artifical Intelligence (IJCAI 2003), pp. 519–526 (August 2003)
Provost, F., Fawcett, T.: Analysis and visualization of classifier performance: comparison under imprecise class and cost distributions. In: Proc. Third Internat. Conf. on Knowledge Discovery and Data Mining (KDD 1997), pp. 43–48. AAAI Press (August 1997)
Fawcett, T.: ROC graphs: notes and practical considerations for data mining. Technical report, HPL-2003-4
Fawcett, T.: An introduction to ROC analysis. Pattern Recogn. 27, 861–874 (2006)
Hérnandez-Orallo, J., Flach, P., Ferri, C.: Brier curves: a new cost-based visualization of classifier performance. In: Proc. Twenty-Eighth Internat. Conf. on Mach. Learn (ICML 2011), pp. 585–592 (June 2011)
Hand, D.J., Till, R.T.: A simple generalization of the area under the ROC curve for multiple class classification problems. Machine Learning 45, 171–186 (2001)
Jain, P., Kapoor, A.: Active learning for large multi-class problems. Comp. Vision and Pattern Recogn (CVPR 2009), 762–769 (June 2009)
Liu, H., et al.: Comparing dissimilarity measures for content-based image retrieval. In: Li, H., Liu, T., Ma, W.-Y., Sakai, T., Wong, K.-F., Zhou, G. (eds.) AIRS 2008. LNCS, vol. 4993, pp. 44–50. Springer, Heidelberg (2008)
Rubner, Y., Tomasi, C., Guibas, L.J.: A metric for distributions with applications to image databases. In: Sixth Internat. Conf. Comp. Vis. (ICCV 1998), pp. 59–66. IEEE Press (January 1998), doi:10.1109/ICCV.19
Raicu, D.S., Varutbangkul, E., Furst, J.D., Armato III, S.G.: Modeling semantics from image data: opportunities from LIDC. Internat. Jour. of Biomed. Eng. and Tech. 3(30:1-2), 83–113 (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Williams, S., Harris, M., Furst, J., Raicu, D. (2013). Area under the Distance Threshold Curve as an Evaluation Measure for Probabilistic Classifiers. In: Perner, P. (eds) Machine Learning and Data Mining in Pattern Recognition. MLDM 2013. Lecture Notes in Computer Science(), vol 7988. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-39712-7_49
Download citation
DOI: https://doi.org/10.1007/978-3-642-39712-7_49
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-39711-0
Online ISBN: 978-3-642-39712-7
eBook Packages: Computer ScienceComputer Science (R0)